Date palm cultivation is a vital agricultural sector in many regions, but it is increasingly threatened by white scale disease, which affects the productivity and quality of the crop. Early and accurate stage-wise detection of this disease is essential for timely intervention and management. This project presents a comprehensive machine learning-based framework that combines both classification and segmentation to automatically detect and localize white scale disease and other visual symptoms (e.g., brown spots) in date palm leaves. The framework incorporates traditional machine learning classifiers such as SVM, KNN, and Random Forest for stage-wise classification, ensemble Method(SVM,KNN,RF), along with a U-Net-based convolutional neural network for semantic segmentation of the infected regions. Extensive data augmentation using Albumentations ensures class balance, while ensemble predictions improve classification robustness. The system also integrates a simple frontend interface with core modules like registration, login, and result viewing for usability. Experimental results demonstrate high accuracy in disease identification and affected region localization, confirming the framework's potential to support agricultural diagnostics and precision farming.

Keywords: Date Palm, White Scale Disease, Machine Learning, Classification, Segmentation, U-Net, Ensemble Learning, Image Processing, Precision Agriculture, Disease Detection